Gas6 deficiency in recipient mice of allogeneic transplantation alleviates hepatic graft-versus-host disease.

Growth arrest-specific gene 6 (Gas6) is expressed in antigen-presenting cells and endothelial cells (ECs) but not in T cells. When wild-type (WT) or Gas6(-/-) mice received allogeneic non-T cell-depleted bone marrow cells, hepatic graft-versus-host disease (GVHD) was alleviated in Gas6(-/-) recipients regardless of donor genotype, but not in WT recipients. T-cell infiltration was more prominent and diffuse in WT than in Gas6(-/-) recipients' liver. When mice received 0.5 x 10(6) allogeneic T cells with T cell-depleted allogeneic bone marrow, clinical signs indicated that GVHD was less severe in Gas6(-/-) than in WT recipients, as shown by a significant improvement of the survival and reduced liver GVHD. These data demonstrate that donor cells were not involved in the protection mechanism. In addition, lack of Gas6 in antigen-presenting cells did not affect WT or Gas6(-/-) T-cell proliferation. We therefore assessed the response of WT or Gas6(-/-) ECs to tumor necrosis factor-alpha. Lymphocyte transmigration was less extensive through Gas6(-/-) than WT ECs and was not accompanied by increases in adhesion molecule levels. Thus, the lack of Gas6 in ECs impaired donor T-cell transmigration into the liver, providing a rationale for considering Gas6 pathway as a potential nonimmunosuppressive target to minimize GVHD in patients receiving allogeneic hematopoietic stem cell transplantation.

Leukocyte-endothelial cell interaction is necessary for photodynamic therapy induced vascular permeabilization.

BACKGROUND AND OBJECTIVE: Photodynamic therapy (PDT) affects vascular barrier function and thus increases vessel permeability. This phenomenon may be exploited to facilitate targeted drug delivery and may lead to a new clinical application of photodynamic therapy. Here, we investigate the role of leukocyte recruitment for PDT-induced vascular permeabilization. STUDY DESIGN/MATERIAL AND METHODS: Fluorescein isothiocyanate dextran (FITC-D, 2,000 kDa) was injected intravenously 120 minutes after focal PDT on striated muscle in nude mice bearing dorsal skinfold chambers (Visudyne® 800 µg/kg, fluence rate 300 mW/cm2 , light dose of 200 J/cm2). Leukocyte interaction with endothelial cells was inhibited by antibodies functionally blocking adhesion molecules ("MABS-PDT" group, n = 5); control animals had PDT but no antibody injection (group "PDT", n = 7). By intravital microscopy, we monitored leukocyte rolling and sticking in real-time before, 90 and 180 minutes after PDT. The extravasation of FITC-D from striated muscle vessels into the interstitial space was determined in vivo during 45 minutes to assess treatment-induced alterations of vascular permeability. RESULTS: PDT significantly increased the recruitment of leukocytes and enhanced the leakage of FITC-D. Neutralization of adhesion molecules before PDT suppressed the rolling of leukocytes along the venular endothelium and significantly reduced the extravasation of FITC-D as compared to control animals (156 ± 27 vs. 11 ± 2 (mean ± SEM, number of WBC/30 seconds mm vessel circumference; P < 0.05) at 90 minutes after PDT and 194 ± 21 vs. 14 ± 4 at 180 minutes after PDT). In contrast, leukocyte sticking was not downregulated by the antibody treatment. CONCLUSION: Leukocyte recruitment plays an essential role in the permeability-enhancing effect of PDT.

Radiotherapy suppresses angiogenesis in mice through TGF-betaRI/ALK5-dependent inhibition of endothelial cell sprouting.

BACKGROUND: Radiotherapy is widely used to treat cancer. While rapidly dividing cancer cells are naturally considered the main target of radiotherapy, emerging evidence indicates that radiotherapy also affects endothelial cell functions, and possibly also their angiogenic capacity. In spite of its clinical relevance, such putative anti-angiogenic effect of radiotherapy has not been thoroughly characterized. We have investigated the effect of ionizing radiation on angiogenesis using in vivo, ex vivo and in vitro experimental models in combination with genetic and pharmacological interventions. PRINCIPAL FINDINGS: Here we show that high doses ionizing radiation locally suppressed VEGF- and FGF-2-induced Matrigel plug angiogenesis in mice in vivo and prevented endothelial cell sprouting from mouse aortic rings following in vivo or ex vivo irradiation. Quiescent human endothelial cells exposed to ionizing radiation in vitro resisted apoptosis, demonstrated reduced sprouting, migration and proliferation capacities, showed enhanced adhesion to matrix proteins, and underwent premature senescence. Irradiation induced the expression of P53 and P21 proteins in endothelial cells, but p53 or p21 deficiency and P21 silencing did not prevent radiation-induced inhibition of sprouting or proliferation. Radiation induced Smad-2 phosphorylation in skin in vivo and in endothelial cells in vitro. Inhibition of the TGF-beta type I receptor ALK5 rescued deficient endothelial cell sprouting and migration but not proliferation in vitro and restored defective Matrigel plug angiogenesis in irradiated mice in vivo. ALK5 inhibition, however, did not rescue deficient proliferation. Notch signaling, known to hinder angiogenesis, was activated by radiation but its inhibition, alone or in combination with ALK5 inhibition, did not rescue suppressed proliferation. CONCLUSIONS: These results demonstrate that irradiation of quiescent endothelial cells suppresses subsequent angiogenesis and that ALK5 is a critical mediator of this suppression. These results extend our understanding of radiotherapy-induced endothelial dysfunctions, relevant to both therapeutic and unwanted effects of radiotherapy.

Population-based genome-wide association studies reveal six loci influencing plasma levels of liver enzymes.

Plasma liver-enzyme tests are widely used in the clinic for the diagnosis of liver diseases and for monitoring the response to drug treatment. There is considerable evidence that human genetic variation influences plasma levels of liver enzymes. However, such genetic variation has not been systematically assessed. In the present study, we performed a genome-wide association study of plasma liver-enzyme levels in three populations (total n = 7715) with replication in three additional cohorts (total n = 4704). We identified two loci influencing plasma levels of alanine-aminotransferase (ALT) (CPN1-ERLIN1-CHUK on chromosome 10 and PNPLA3-SAMM50 on chromosome 22), one locus influencing gamma-glutamyl transferase (GGT) levels (HNF1A on chromosome 12), and three loci for alkaline phosphatase (ALP) levels (ALPL on chromosome 1, GPLD1 on chromosome 6, and JMJD1C-REEP3 on chromosome 10). In addition, we confirmed the associations between the GGT1 locus and GGT levels and between the ABO locus and ALP levels. None of the ALP-associated SNPs were associated with other liver tests, suggesting intestine and/or bone specificity. The mechanisms underlying the associations may involve cis- or trans-transcriptional effects (some of the identified variants were associated with mRNA transcription in human liver or lymphoblastoid cells), dysfunction of the encoded proteins (caused by missense variations at the functional domains), or other unknown pathways. These findings may help in the interpretation of liver-enzyme tests and provide candidate genes for liver diseases of viral, metabolic, autoimmune, or toxic origin. The specific associations with ALP levels may point to genes for bone or intestinal diseases.

The fibrinogen- and fibronectin-binding domains of Staphylococcus aureus fibronectin-binding protein A synergistically promote endothelial invasion and experimental endocarditis.

Staphylococcus aureus experimental endocarditis relies on sequential fibrinogen binding (for valve colonization) and fibronectin binding (for endothelial invasion) conferred by peptidoglycan-attached adhesins. Fibronectin-binding protein A (FnBPA) reconciles these two properties--as well as elastin binding--and promotes experimental endocarditis by itself. Here we attempted to delineate the minimal subdomain of FnBPA responsible for fibrinogen and fibronectin binding, cell invasion, and in vivo endocarditis. A large library of truncated constructs of FnBPA was expressed in Lactococcus lactis and tested in vitro and in animals. A 127-amino-acid subdomain spanning the hinge of the FnBPA fibrinogen-binding and fibronectin-binding regions appeared necessary and sufficient to confer the sum of these properties. Competition with synthetic peptides could not delineate specific fibrinogen- and fibronectin-binding sites, suggesting that dual binding arose from protein folding, irrespective of clearly defined binding domains. Moreover, coexpressing the 127-amino-acid subdomain with remote domains of FnBPA further increased fibrinogen binding by &gt; or =10 times, confirming the importance of domain interactions for binding efficacy. In animals, fibrinogen binding (but not fibronectin binding) was significantly associated with endocarditis induction, whereas both fibrinogen binding and fibronectin binding were associated with disease severity. Moreover, fibrinogen binding also combined with fibronectin binding to synergize the invasion of cultured cell lines significantly, a feature correlating with endocarditis severity. Thus, while fibrinogen binding and fibronectin binding were believed to act sequentially in colonization and invasion, they appeared unexpectedly intertwined in terms of both functional anatomy and pathogenicity (in endocarditis). This unforeseen FnBPA subtlety might bear importance for the development of antiadhesin strategies.

Physical training and hypertension have opposite effects on endothelial brain-derived neurotrophic factor expression.

AIMS: Changes in circulating brain-derived neurotrophic factor (BDNF) levels were reported in patients with or at risk for cardiovascular diseases associated with endothelial dysfunction, suggesting a link between BDNF and endothelial functionality. However, little is known on cardiovascular BDNF. Our aim was to investigate levels/localization, function, and relevance of cardiovascular BDNF. METHODS AND RESULTS: BDNF levels (western blotting) and localization (immunostaining) were assessed in the heart and aorta from rats with impaired (spontaneously hypertensive rats [SHR]), normal (Wistar Kyoto rats [WKY]), and improved (SHR and WKY subjected to physical training) endothelial function. BDNF levels were also measured in cultured endothelial cells (CECs) subjected to low and high shear stress. The cardiovascular effects of BDNF were investigated in isolated aortic rings and hearts. The results showed high BDNF levels in the heart and aorta, the expression being prominent in endothelial cells as compared with other cell types. Exogenous BDNF vasodilated aortic rings but changed neither coronary flow nor cardiac contractility. Hypertension was associated with decreased expression of BDNF in the endothelium, whereas physical training led to endothelial BDNF up-regulation not only in WKY but also in SHR. Exposure of CECs to high shear stress stimulated BDNF production and secretion. CONCLUSION: Cardiovascular BDNF is mainly localized within endothelial cells in which its expression is dependent on endothelial function. These results open new perspectives on the role of endothelial BDNF in cardiovascular health.

mTORC2 regulates PGE(2)-mediated endothelial cell survival and migration

Prostaglandin E-2 (PGE(2)) promotes angiogenesis by in part inducing endothelial cell survival and migration. The present study examined the role of mTOR and its two complexes, mTORC1 and mTORC2, in PGE(2)-mediated endothelial cell responses. We used small interfering RNA (siRNA) to raptor or rictor to block mTORC1 or mTORC2, respectively. We observed that down-regulation of mTORC2 but not mTORC1 reduced baseline and PGE(2)-induced endothelial cell survival and migration. At the molecular level, we found that knockdown of mTORC2 inhibited PGE2-mediated Rac and Akt activation two important signaling intermediaries in endothelial cell migration and survival, respectively. In addition, inhibition of mTORC2 by prolonged exposure of endothelial cells to rapamycin also prevented PGE2-mediated endothelial cell survival and migration confirming the results obtained with the siRNA approach. Taken together these results show that mTORC2 but not mTORC1 is an important signaling intermediary in PGE2-mediated endothelial cell responses.

Expressed isolated integrin beta1 subunit cytodomain induces endothelial cell death secondary to detachment.

Expression of isolated beta integrin cytoplasmic domains in cultured endothelial cells was reported to induce cell detachment and death. To test whether cell death was the cause or the consequence of cell detachment, we expressed isolated integrin beta1 cytoplasmic and transmembrane domains (CH1) in cultured human umbilical vein endothelial cells (HUVEC), and monitored detachment, viability, caspase activation and signaling. CH1 expression induced dose-dependent cell detachment. At 24 h over 90% of CH1-expressing HUVEC were detached but largely viable (>85%). No evidence of pro-caspase-8,-3, and PARP cleavage or suppression of phosphorylation of ERK, PKB and Ikappa-B was observed. The caspase inhibitor z-VAD did not prevent cell detachment. At 48 h, however, CH1-expressing cells were over 50% dead. As a comparison trypsin-mediated detachment resulted in a time-dependent cell death, paralleled by caspase-3 activation and suppression of ERK, PKB and Ikappa-B phosphoyrylation at 24 h or later after detachment. HUVEC stimulation with agents that strengthen integrin-mediated adhesion (i.e. PMA, the Src inhibitor PP2 and COMP-Ang1) did not prevent CH1-induced detachment. Expression of CH1 in rat carotid artery endothelial cells in vivo caused endothelial cell detachment and increased nuclear DNA fragmentation among detached cells. A construct lacking the integrin cytoplasmic domain (CH2) had no effect on adhesion and cell viability in vitro and in vivo. These results demonstrate that isolated beta1 cytoplasmic domain expression induces caspase-independent detachment of viable endothelial cells and that death is secondary to detachment (i.e. anoikis). They also reveal an essential role for integrins in the adhesion and survival of quiescent endothelial cells in vivo.

Caveolin-1 opens endothelial cell junctions by targeting catenins.

AIMS: A fundamental phenomenon in inflammation is the loss of endothelial barrier function, in which the opening of endothelial cell junctions plays a central role. However, the molecular mechanisms that ultimately open the cell junctions are largely unknown.¦METHODS AND RESULTS: Impedance spectroscopy, biochemistry, and morphology were used to investigate the role of caveolin-1 in the regulation of thrombin-induced opening of cell junctions in cultured human and mouse endothelial cells. Here, we demonstrate that the vascular endothelial (VE) cadherin/catenin complex targets caveolin-1 to endothelial cell junctions. Association of caveolin-1 with VE-cadherin/catenin complexes is essential for the barrier function decrease in response to the pro-inflammatory mediator thrombin, which causes a reorganization of the complex in a rope ladder-like pattern accompanied by a loss of junction-associated actin filaments. Mechanistically, we show that in response to thrombin stimulation the protease-activated receptor 1 (PAR-1) causes phosphorylation of caveolin-1, which increasingly associates with β- and γ-catenin. Consequently, the association of β- and γ-catenin with VE-cadherin is weakened, thus allowing junction reorganization and a decrease in barrier function. Thrombin-induced opening of cell junctions is lost in caveolin-1-knockout endothelial cells and after expression of a Y/F-caveolin-1 mutant but is completely reconstituted after expression of wild-type caveolin-1.¦CONCLUSION: Our results highlight the pivotal role of caveolin-1 in VE-cadherin-mediated cell adhesion via catenins and, in turn, in barrier function regulation.

Bone marrow mesenchymal stem cells stabilize already-formed aortic aneurysms more efficiently than vascular smooth muscle cells in a rat model.

PURPOSE: Abdominal aortic aneurysms (AAAs) expand because of aortic wall destruction. Enrichment in Vascular Smooth Muscle Cells (VSMCs) stabilizes expanding AAAs in rats. Mesenchymal Stem Cells (MSCs) can differentiate into VSMCs. We have tested the hypothesis that bone marrow-derived MSCs (BM-MSCs) stabilizes AAAs in a rat model. MATERIAL AND METHODS: Rat Fischer 344 BM-MSCs were isolated by plastic adhesion and seeded endovascularly in experimental AAAs using xenograft obtained from guinea pig. Culture medium without cells was used as control group. The main criteria was the variation of the aortic diameter at one week and four weeks. We evaluated the impact of cells seeding on inflammatory response by immunohistochemistry combined with RT-PCR on MMP9 and TIMP1 at one week. We evaluated the healing process by immunohistochemistry at 4 weeks. RESULTS: The endovascular seeding of BM-MSCs decreased AAA diameter expansion more powerfully than VSMCs or culture medium infusion (6.5% ± 9.7, 25.5% ± 17.2 and 53.4% ± 14.4; p = .007, respectively). This result was sustained at 4 weeks. BM-MSCs decreased expression of MMP-9 and infiltration by macrophages (4.7 ± 2.3 vs. 14.6 ± 6.4 mm(2) respectively; p = .015), increased Tissue Inhibitor Metallo Proteinase-1 (TIMP-1), compared to culture medium infusion. BM-MSCs induced formation of a neo-aortic tissue rich in SM-alpha active positive cells (22.2 ± 2.7 vs. 115.6 ± 30.4 cells/surface units, p = .007) surrounded by a dense collagen and elastin network covered by luminal endothelial cells. CONCLUSIONS: We have shown in this rat model of AAA that BM-MSCs exert a specialized function in arterial regeneration that transcends that of mature mesenchymal cells. Our observation identifies a population of cells easy to isolate and to expand for therapeutic interventions based on catheter-driven cell therapy.

Histones from dying renal cells aggravate kidney injury via TLR2 and TLR4.

In AKI, dying renal cells release intracellular molecules that stimulate immune cells to secrete proinflammatory cytokines, which trigger leukocyte recruitment and renal inflammation. Whether the release of histones, specifically, from dying cells contributes to the inflammation of AKI is unknown. In this study, we found that dying tubular epithelial cells released histones into the extracellular space, which directly interacted with Toll-like receptor (TLR)-2 (TLR2) and TLR4 to induce MyD88, NF-κB, and mitogen activated protein kinase signaling. Extracellular histones also had directly toxic effects on renal endothelial cells and tubular epithelial cells in vitro. In addition, direct injection of histones into the renal arteries of mice demonstrated that histones induce leukocyte recruitment, microvascular vascular leakage, renal inflammation, and structural features of AKI in a TLR2/TLR4-dependent manner. Antihistone IgG, which neutralizes the immunostimulatory effects of histones, suppressed intrarenal inflammation, neutrophil infiltration, and tubular cell necrosis and improved excretory renal function. In summary, the release of histones from dying cells aggravates AKI via both its direct toxicity to renal cells and its proinflammatory effects. Because the induction of proinflammatory cytokines in dendritic cells requires TLR2 and TLR4, these results support the concept that renal damage triggers an innate immune response, which contributes to the pathogenesis of AKI.

Isolation of cardiovascular precursor cells from the human fetal heart.

Weakening of cardiac function in patients with heart failure results from a loss of cardiomyocytes in the damaged heart. Cell replacement therapies as a way to induce myocardial regeneration in humans could represent attractive alternatives to classical drug-based approaches. However, a suitable source of precursor cells, which could produce a functional myocardium after transplantation, remains to be identified. In the present study, we isolated cardiovascular precursor cells from ventricles of human fetal hearts at 12 weeks of gestation. These cells expressed Nkx2.5 but not late cardiac markers such as α-actinin and troponin I. In addition, proliferating cells expressed the mesenchymal stem cell markers CD73, CD90, and CD105. Evidence for functional cardiogenic differentiation in vitro was demonstrated by the upregulation of cardiac gene expression as well as the appearance of cells with organized sarcomeric structures. Importantly, differentiated cells presented spontaneous and triggered calcium signals. Differentiation into smooth muscle cells was also detected. In contrast, precursor cells did not produce endothelial cells. The engraftment and differentiation capacity of green fluorescent protein (GFP)-labeled cardiac precursor cells were then tested in vivo after transfer into the heart of immunodeficient severe combined immunodeficient mice. Engrafted human cells were readily detected in the mouse myocardium. These cells retained their cardiac commitment and differentiated into α-actinin-positive cardiomyocytes. Expression of connexin-43 at the interface between GFP-labeled and endogenous cardiomyocytes indicated that precursor-derived cells connected to the mouse myocardium. Together, these results suggest that human ventricular nonmyocyte cells isolated from fetal hearts represent a suitable source of precursors for cell replacement therapies.

Extracellular vesicles from human cardiac progenitor cells inhibit cardiomyocyte apoptosis and improve cardiac function after myocardial infarction.

AIMS: Recent evidence suggests that cardiac progenitor cells (CPCs) may improve cardiac function after injury. The underlying mechanisms are indirect, but their mediators remain unidentified. Exosomes and other secreted membrane vesicles, hereafter collectively referred to as extracellular vesicles (EVs), act as paracrine signalling mediators. Here, we report that EVs secreted by human CPCs are crucial cardioprotective agents. METHODS AND RESULTS: CPCs were derived from atrial appendage explants from patients who underwent heart valve surgery. CPC-conditioned medium (CM) inhibited apoptosis in mouse HL-1 cardiomyocytic cells, while enhancing tube formation in human umbilical vein endothelial cells. These effects were abrogated by depleting CM of EVs. They were reproduced by EVs secreted by CPCs, but not by those secreted by human dermal fibroblasts. Transmission electron microscopy and nanoparticle tracking analysis showed most EVs to be 30-90 nm in diameter, the size of exosomes, although smaller and larger vesicles were also present. MicroRNAs most highly enriched in EVs secreted by CPCs compared with fibroblasts included miR-210, miR-132, and miR-146a-3p. miR-210 down-regulated its known targets, ephrin A3 and PTP1b, inhibiting apoptosis in cardiomyocytic cells. miR-132 down-regulated its target, RasGAP-p120, enhancing tube formation in endothelial cells. Infarcted hearts injected with EVs from CPCs, but not from fibroblasts, exhibited less cardiomyocyte apoptosis, enhanced angiogenesis, and improved LV ejection fraction (0.8 ± 6.8 vs. -21.3 ± 4.5%; P < 0.05) compared with those injected with control medium. CONCLUSION: EVs are the active component of the paracrine secretion by human CPCs. As a cell-free approach, EVs could circumvent many of the limitations of cell transplantation.

Distinctive role of integrin-mediated adhesion in TNF-induced PKB/Akt and NF-kappaB activation and endothelial cell survival.

Tumor necrosis factor (TNF) is a pro-inflammatory cytokine exerting pleiotropic effects on endothelial cells. Depending on the vascular context it can induce endothelial cell activation and survival or death. The microenvironmental cues determining whether endothelial cells will survive or die, however, have remained elusive. Here we report that integrin ligation acts permissive for TNF-induced protein kinase B (PKB/Akt) but not nuclear factor (NF)-kappaB activation. Concomitant activation of PKB/Akt and NF-kappaB is essential for the survival of endothelial cells exposed to TNF. Active PKB/Akt strengthens integrin-dependent endothelial cell adhesion, whereas disruption of actin stress fibers abolishes the protective effect of PKB/Akt. Integrin-mediated adhesion also represses TNF-induced JNK activation, but JNK activity is not required for cell death. The alphaVbeta3/alphaVbeta5 integrin inhibitor EMD121974 sensitizes endothelial cells to TNF-dependent cytotoxicity and active PKB/Akt attenuates this effect. Interferon gamma synergistically enhanced TNF-induced endothelial cell death in all conditions tested. Taken together, these observations reveal a novel permissive role for integrins in TNF-induced PKB/Akt activation and prevention of TNF-induced death distinct of NF-kappaB, and implicate the actin cytoskeleton in PKB/Akt-mediated cell survival. The sensitizing effect of EMD121974 on TNF cytotoxicity may open new perspectives to the therapeutic use of TNF as anticancer agent.

mTORC2 is an important signaling intermediary in VEGF-mediated endothelial cell proliferation, survival and migration

Objective: The vascular endothelial growth factor (VEGF) is a prominent¦contributor of tumor angiogenesis. VEGF induces endothelial cell migration,¦proliferation and survival, which are critical steps for the development of new¦blood vessels, through the activation of the Mek/Erk and PI3K/Akt signaling¦pathways. Recent findings have demonstrated that mTORC2 regulates Akt and¦Erk in endothelial cells. The role of mTORC2 in VEGF-mediated endothelial¦cell responses has however not been characterized.¦Methods: We used human umbilical vein endothelial cells (HUVEC). The¦effects of VEGF on the Mek/Erk and PI3K/Akt pathway were analyzed by¦Western blot. Inhibition of mTORC2 was achieved using small interfering¦RNAs to rictor. Cell proliferation rate was assessed by BrdU incorporation and¦immunocytofluorescence. Apoptosis rate was determined by ELISA as well as¦propidium iodine staining and FACS analysis. Migration of endothelial cells¦was evaluated using a modified Boyden chamber assay.¦Results:Wefound thatVEGF activatesmTORC2 in endothelial cells. Indeed,¦treatment of endothelial cells with VEGF increases Akt phosphorylation, a¦downstream effector of mTORC2. We have further determined the role¦of mTORC2 in VEGF signaling by knocking down rictor, a component¦of mTORC2. We observed that VEGF failed to activate Akt and Erk in¦endothelial cells transfected with rictor siRNA. To next analyze the functional¦significance of mTORC2 inhibition on VEGF-mediated endothelial cell¦responses we performed proliferation, survival and migration assays. We found¦that VEGF failed to induce endothelial cell proliferation, survival and migration¦in endothelial cell lacking mTORC2 activity.¦Conclusion: These results show that mTORC2 is an important signaling¦intermediary in VEGF-induced endothelial cell responses and thus represents¦an interesting target to block VEGF-induced angiogenesis.